68Ga-FAPI-04 vs. 18F-FDG in a longitudinal preclinical PET imaging of metastatic breast cancer

This longitudinal study aims to evaluate the performance of 68 Ga-FAPI-04 and 18F-FDG and to profile the dynamic process of tumor metastasis in a preclinical 4T1 breast cancer model. Although both of these two radioligands are wildly used in clinic, no study was reported on their performance in the longitudinal monitoring of tumor metastasis. Also, no correlation between the expression level of fibroblast activation protein (FAP) and the development of tumor metastasis has been elucidated previously. In this study, we evaluated the performance of 68 Ga-FAPI-04 and 18F-FDG PET during the entire process of tumor metastasis, and their potential for the early diagnosis of tumor metastasis. We also clarified the correlation of uptakes as well as the signal-to-background (S/B) ratios between these two probes at different stages of tumor metastasis. Forty 4T1 metastatic breast cancer murine models were established using female BALB/c mice, followed by the longitudinal imaging with 68 Ga-FAPI-04 and 18F-FDG once a week for up to 6 weeks. In vitro hematoxylin and eosin (H&E) and immunochemistry (IHE) staining were performed to evaluate FAP expression on the metastatic lesions. Further statistical analysis was performed to evaluate the correlation of 68 Ga-FAPI-04 and 18F-FDG uptake (%ID/cc) at different stages of the metastasis. 68 Ga-FPAI-04 holds an advantage over 18F-FDG with higher sensitivity at the early stage of tumor metastasis. However, with the progress of tumor metastasis, uptake of 68 Ga-FAPI-04 decreases and becomes less sensitive than 18F-FDG. There is also no direct correlation between uptake or S/B ratios of 68 Ga-FAPI-04 and 18F-FDG during this dynamic process. 68 Ga-FAPI-04 is more sensitive than 18F-FDG in detecting the early stage of tumor metastasis, but becomes less sensitive than 18F-FDG at the late stage of tumor metastasis. We envision this result would be meaningful for the explanation of the 68 Ga-FAPI-04 and 18F-FDG imaging both in the future clinic and preclinic studies.


Introduction
Tumor metastasis detection is crucial for both cancer diagnosis and treatment. In clinic, tumor metastasis is normally detected using 2-deoxy-2-[ 18 F]fluoroglucose ( 18 F-FDG) PET imaging. As an imaging agent to profile tumor metabolism, 18 F-FDG is used directly to visualize tumor cells instead of tumor microenvironment. 68 Ga-FAPI-04, a novel radioligand to target fibroblast activation protein (FAP), was recently employed to image tumor microenvironment, especially for cancer associate fibroblasts (CAFs) with promising imaging result [1][2][3].
CAFs are generally believed to be one of the main components in the extracellular matrix to promote cancer [4][5][6], even before the cancer cells spread to the metastasis, secretion of exosomes [7] from original tumor mass can raise Fan Ding and Chen Huang contributed equally to this work This article is part of the Topical Collection on Preclinical Imaging * Jianjun Liu nuclearj@163.com * Dewei Tang acetdw@126.com and activate CAFs cell for pre-metastatic niche [8,9]. CAFs are essential for tumor initiation, progression, and metastasis and it normally forms at the early stage of the tumor metastasis [10], which is regarded to be prior to the tumor cell migration or colonization as shown in Fig. 1. The early stage of metastasis cannot be visualized with 18 F-FDG PET imaging, since little or tiny amount of tumor cells exist in the niche as shown in Fig. 1. However, it can be visualized by 68 Ga-FAPI-04 PET imaging as for the large amount of CAFs and FAP expressed in the niche [7]. Thus, we envision 68 Ga-FAPI-04 imaging holds greater promise and higher sensitivity in detecting the early stage of metastasis when compared to 18 F-FDG imaging. As for components of extracellular matrix and CAFs may change dynamically during tumor growth and metastasis [11,12], the uptake of 68 Ga-FAPI-04 may also features a dramatic change during this dynamic process, which will definitely affect its detection rate in clinic. In order to fully decipher this dynamic process, we employed longitudinal 68 Ga-FAPI-04 imaging to evaluate the variation of CAFs and FAP expression. We also elucidated the relationship between the uptake and signal-tobackground (S/B) ratio of 68 Ga-FAPI-04 and 18 F-FDG with the aim to clarify the relationship between FAP expression and the growth state of metastatic lesions. In clinic studies, it is still difficult to perform the longitudinal 68 Ga-FAPI-04 and 18 F-FDG imaging of patients using a series of PET scans. Clinic studies at present can only be performed to evaluate the performance of 18 F-FDG and   68 Ga-FAPI-04 at certain stages of tumor metastasis [13][14][15][16][17][18][19][20]. 68 Ga-FAPI-04 seems to be more sensitive than 18 F-FDG, whether in detecting brain metastasis of lung adenocarcinoma [21], gastrointestinal tumors [16], or bone and peritoneum metastasis [22,23]. 68 Ga-FAPI-04 also demonstrated a clearer tumor delineation, and a higher tumor-to-background contrast [3]. However, in a study with lymph node metastasis of head and neck tumor, Serfling et al. observed that 68 Ga-FAPI-04 was not as sensitive as 18 F-FDG (47% vs. 82%) [24]. Qin et al. also reported that 68 Ga-FAPI-04 outperforms 18 F-FDG in delineating the primary tumor and detecting suspected distant metastases, but cannot detected as many positive lymph nodes as 18 F-FDG (48 vs 100) [20]. In addition, by examining stromal FAP in patients with colorectal cancer and its correlation with clinical parameters, Henry et al. found stromal FAP intensity and semiquantitative analysis were inversely correlated to cancer stage and sizes of xenografted tumors. They considered FAP pathways may predominate early in the course of smaller tumors to facilitate tumor invasion and tumor motility, but may be diluted and diminished as other growth promoting pathways predominate [25]. Since different stages of tumor metastasis may feature different FAP expression profiles and thus 68 Ga-FAPI-04 uptakes, so far it is still hard to determine whether 68 Ga-FAPI-04 is superior to 18 F-FDG for the entire tumor CAFs and FAP are crucial component for this niche formation metastatic process, or at which certain stage 68 Ga-FAPI-04 is more sensitive than 18 F-FDG.
Considering this, we performed this longitudinal study using 68 Ga-FAPI-04 and 18 F-FDG to profile the dynamic change of tumor uptakes with 4T1 metastatic breast cancer murine model. This study also aims to fully evaluate their potential for the diagnosis of tumor metastasis, as well as to clarify the correlation between uptakes of these two radioligands at different tumor metastatic stages.

Experimental protocol
As shown in Fig. 2, we established forty 4T1 metastatic breast cancer murine models and performed longitudinal PET imaging for up to 6 weeks, which monitored the entire process of metastasis from breast to lungs and lymph nodes. In detail, imaging studies were performed one week after tumor cell injection, with mice imaged by 68 Ga-FAPI-04 first and then by 18 F-FDG on the following day. All mice were sacrificed and the tissues were collected for in vitro immunohistochemical analysis.

Murine model of metastatic breast cancer
All experiments were performed in vivo abiding by the guidelines of the Animal Ethics Committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong University. BALB/c female mice (six-week old, n = 40) were purchased from Obio Technology (Shanghai, China). All mice received standard care, including free access to food and water, a 12/12 light/dark cycle, appropriate temperature and humidity. We established the 4T1 metastatic breast cancer model by orthotopic injecting of 4T1 cancer metastatic cells (1 × 10 6 cells in the incompleted RPMI-1640 medium) into the seventh mammary fat pads. The primary tumors were resected at week 2 for all the animal models in this study [26].

Radiolabeling
Isotope Ga-68 was eluted from 68 Ge/ 68 Ga generator by 5.0 ml of 0.1 M HCl. In this study, 2.5 ml eluent containing the highest activity (333.0-370.0 MBq) was used for the following radiolabeling. The Ga-68 eluent was transferred into a reactor vial containing 20.0 µg of DOTA-FAPI-04 (purchased from OptimLib, Shanghai) mixed with NaOAc (78.1 µmol) to adjust the pH of the reaction mixture to 3.5, then heated at 100 °C for 10 min. After cooling, the reaction mixture was passed through a C18 Sep-Pak Light solidphase extraction cartridge for further purification, which was preconditioned by purging with ethanol (5.0 mL) and water (10.0 mL), 68 Ga-FAPI-04 was eluted from the cartridge with 2.0 mL of ethanol/water (1/1, v/v), the product was then analyzed by radioTLC for quality control with radiochemical purity more than 95%.

PET/CT imaging and analysis
68 Ga-FPAI-04 PET imaging, followed by 18 F-FDG PET imaging, was performed using MicroPET/CT system (Inviscan, France) once a week after 4T1 cell injection (Fig. 2). The mice were first anesthetized by inhalation of 2.0% isoflurane (Abbott Laboratories, Chicago, USA), and then imaged with a 30 min dynamic PET/CT scan. At the beginning of this scan, 3.7-7.4 MBq of 68 Ga-FAPI-04 in 200.0 µl saline was injected intravenously via the tail-vein for each animal. 18 F-FDG PET/CT scan was performed on the following day using the same cohort of mice fasted overnight before the 18 F-FDG injection. During this scan, mice were kept warm on a heating pad and anesthetized by inhalation of 2.0% isoflurane to reduce their activity for low muscle uptake. 18 68 Ga-FAPI-04 imaging was performed first and then followed by 18 F-FDG PET/CT scan. We resected all primary tumors at week 2 and all mice were sacrificed and organs were collected for H&E and immunohistochemistry (IHC) staining after the dynamic scan at week 5 and week 6 was injected via the tail vein, followed by a 1 h uptake period and a 10 min static PET scan. PET list-mode data was acquired using a γ-ray energy window of 350-650 keV with a coincidence-timing window of 3.432 ns. CT Cone beam reconstruction resulted in 192 × 192 matrix and PET reconstruction utilized OSEM-3D, 4 iterations, resulting in a 256 × 256 matrix. All PET images were analyzed using the Inveon Research Workplace (Siemens, Knoxville, USA). Tissue uptakes were quantified by percentage injected dose per cubic centimeter (%ID/cc). Signal-to-background (S/B) ratio was calculated using the uptake of the targeted lesions divided by the uptake of the upper limb muscles for each mouse.
Volumes-of-interest (VOIs) analyses were performed in this study to profile the tumor uptake of 68 Ga-FAPI-04 and 18 F-FDG at different stages of lung metastasis. In order to monitor the variation of radiotracers uptake longitudinally, PET images from the same mouse (from Week 2 to Week 6) with different radiotracers were all co-registered. In detail, using Inveon Research Workplace 4.0 (Siemens Medical Solutions USA, Knoxville, TN, USA), corresponding CT images were first co-registered to make the spatial alignment of lung and rib regions. The transformation matrix for this co-registration was generated and then treated as an input to register the corresponding PET images. With these registered PET images, 68 Ga-FAPI-04 positive sphere VOIs (normally appears at week 4) were visually selected and manually drawn. Using the same sphere VOIs, corresponding uptakes of 18 F-FDG or 68 Ga-FAPI-04 was then calculated for PET images obtained from week 2 to week 6. The VOIs featuring visually positive 68 Ga-FAPI-04 or 18 F-FDG uptakes were also double checked by S/B ratios to confirm whether the positive VOIs feature significantly higher S/B ratios than the ones obtained from the normal lung tissues during week 2 (both 68 Ga-FAPI-04 or 18 F-FDG negative). If not, they are regarded as negative 18 F-FDG or 68 Ga-FAPI-04 VOIs. All ID%/cc were calculated as mean uptake in this study.

Histological analysis
All the mice were sacrificed and tissues were harvested and then fixed in 10% paraformaldehyde. Series of 10.0 µm sections were sliced and paraffin-embedded for staining of mice tissues from week 5 (n = 3) and week 6 (n = 6). The hematoxylin and eosin (H&E) staining were performed using the tissue slices with high uptake of either 18 F-FDG or 68 Ga-FAPI-04 during imaging. Additionally, immunohistochemistry (IHC) staining was performed to characterize FAP expression with tissue slices from lung and lymph metastasis. For immunohistochemical examination, all sections were bathed in ethylenediaminetetraacetic acid (EDTA) buffer for antigen retrieval; then, sections were blocked with 5.0% bovine serum albumin (BSA) and 2.0% horse serum. Slices were washed with PBS and overnight incubation at 4.0 °C with anti-rabbit FAP antibody (Abcam, USA). After washed with PBS, slices were incubated with goat anti-mouse IgG (Abcam, USA) and DAB (3,3′-diaminobenzidine solution), and counterstaining was performed with hematoxylin. Confocal images were acquired using a Nikon A1R confocal microscope. The % area of FAP positive for the metastatic lesions from anti-FAP IHC was analyzed for both week 5 and 6 using ImageJ.

Western blot analysis
To obtain total protein lysates, fresh metastasis niche was ground to powder by adding lipid nitrogen. Grinding tissue were lysed using cell lysates containing mixed proteinase inhibitors. The lysate was incubated on ice for 30 min followed by centrifugation at 4 °C 15000 g for 30 min. The protein concentration of each sample was assayed using the bicinchoninic acid (BCA) assay. The cytoplasmic and nuclear extracts were separated by extraction regents (Invitrogen, USA). Samples were loaded on 7.5% or 10% SDS-PAGE. After electrophoresis, the proteins were transferred from gels to PVDF membranes. The membranes were blocked in 5% low-fat dried milk in TBST for an hour at room temperature and then incubated with the primary antibody (anti-FAP antibody (Abcam, USA)) overnight at 4 °C. The immunoreactive bands were visualized by the ECL Plus system (Tanon, China).

Statistical analysis
Statistical analysis was performed using GraphPad Prism and Microsoft Excel. Semiquantitative analysis of immunohistochemistry was performed using ImageJ. Results are presented as the mean value ± standard error of the mean (SEM) if not stated otherwise. Group data were compared using the two-tailed Student t test, and correlation between 68 Ga-FAPI-04 and 18 F-FDG uptakes were analyzed using Pearson's rank correlation. p < 0.05 was considered statistically significant.

Results
In total, 40 female BALB/c mice was established for the longitudinal imaging with 24 mice imaged during the entire process and 16 mice died or sacrificed in the middle of this study (Fig. 2). The overall detection rate of pulmonary metastasis by both 18 F-FDG and 68 Ga-FAPI-04 were 50.0% (12 vs. 24). The other 12 mice cannot be detected by either 18 F-FDG or 68 Ga-FAPI-04, with S/B ratio close to 1.0.

Comparison of 68 Ga-FPAI and 18 F-FDG in pulmonary metastasis
As shown in Fig. 3a, during the development of metastasis, both of these two radioligands did not demonstrate a high detection rate before week 4. During week 5, 68 Ga-FAPI-04 showed an increased signal-to-background (S/B) ratio (close to 3.0), while at the same position 18 F-FDG still did not demonstrate a high detection rate with a S/B ratio close to 1.0. However, at week 6, 68 Ga-FPAI-04 uptake decreases and the S/B ratio drops to one, while the uptake of 18 F-FDG increases and the S/B ratio increases up to 4.0. After the 18 F-FDG PET scan, the mice were sacrificed and the region with high uptake of 18 F-FDG was analyzed with IHC staining. As shown in Fig. 3c, H&E stain of this region demonstrated a large metastatic lesion, with the location and the size corresponding to the metastatic lesion detected using 68 Ga-FAPI-04 at week 5 and 18 F-FDG at week 6. As expected, IHC assay showed little FAP expression both in the tumor and the tumor-adjacent tissue during week 6 ( Fig. 3c). This low-level expression of FAP is also corresponding to the low uptake of 68 Ga-FAPI-04 during week 6 as shown in Fig. 3a. In order to fully profile this dynamic process of metastasis, further in vitro IHC and H&E analysis was also performed to evaluate FAP expression on metastatic lesion at week 5, which belongs to an earlier stage of metastasis when compared to week 6. Interestingly, we observed a high expression of FAP in the tiny metastatic lesions as shown in Fig. 3b. The high expression of FAP is also corresponding to a high uptake of 68 Ga-FAPI-04 visualized on the imaging at week 5. Interestingly, Fig. 3b and c show a significant difference of FAP expression between week 5 and week 6, indicating the dynamic change of the FAP expression during the metastasis.
Overall, in this longitudinal study, the uptake of 68 Ga-FAPI-04 increased dramatically from week 4 to week 5 (1.417 ± 0.082 vs. 2.158 ± 0.142), but decreased dramatically from week 5 to week 6 (2.158 ± 0.142 vs. 1.658 ± 0.093) (Fig. 3d). While for 18 F-FDG, the uptake did not change from week 4 to week 5(2.142 ± 0.091 vs. 2.508 ± 0.146), but increased dramatically from week 5 to week 6 (2.508 ± 0.146 vs. 3.658 ± 0.239). This dynamic change also influences the detection rate for both radioligands as shown in Fig. 3e. In detail, we noticed a higher detection rate for 68 Ga-FAPI-04 at week 5 (12 vs. 2), but a lower detection rate at week 6 (1 vs. 12), when compared to 18 F-FDG (Fig. 3e). Figure 3f demonstrates the changes in the proportion of FAP positive areas between week 5 (16.52 ± 1.32) and week 6 (3.02 ± 0.37). The proportion of FAP positive area was positively correlated with 68 Ga-FAPI-04 uptakes as shown in Fig. 3g. Accordingly, western blot analysis also demonstrates a higher FAP expression at the metastatic niche for week 5 when compared with week 6 as shown in Fig. 3h. The overall change of 18 F-FDG and 68 Ga-FAPI-04 uptakes from a cohort of mice (n = 12) is also demonstrated in Fig. 4a, showing 18 F-FDG uptake increases from week 3 to week 6 gradually. Contrary to 18 F-FDG, uptake of 68 Ga-FAPI-04 increases from week 3 to week 5 gradually, but drops from week 5 to week 6. The dynamic change of S/B ratio between 68 Ga-FAPI-04 and 18 F-FDG is also shown in Fig. 4b, reflecting the highest S/B ratio of 68 Ga-FAPI-04 was observed at week 5 (2.14 ± 0.16) and the highest S/B ratio of 18 F-FDG at week 6 (1.65 ± 0.097). No significant difference was found between the uptakes of 68 Ga-FAPI-04 at week 3 and week 4 (p > 0.05). Furthermore, the significant difference was also found between the uptakes of 68 Ga-FAPI-04 at week 4 and week 5 (p < 0.0001), as well as between week 5 and week 6 (p < 0.001). Similarly, no significant difference was found between the uptakes of 18 F-FDG at week 3 and week 4 (p > 0.05), as well as between week 4 and week 5 (p > 0.05). But the significant difference was observed between week 5 and week 6 (p < 0.0001). For the S/B ratio of 18 F-FDG, no significant difference was observed between week 3 and week 4 (p > 0.05), as well as week 4 and week 5 (p > 0.05). But significant difference was observed between week 5 and week 6 (p < 0.0001). For the S/B ratio of 68 Ga-FAPI-04, no significant difference was observed between week 3 and week 4 (p > 0.05), but significant difference was observed between week 4 and week 5 (p < 0.005), as well as week 5 and week 6 (p < 0.005). The S/B ratio in Fig. 4b also indicates 68 Ga-FAPI-04 is more sensitive than 18 F-FDG from week 3 to week 5 (2.138 ± 0.162 vs. 1.174 ± 0.073), but turns to be less sensitive from week 5 to week 6 (1.459 ± 0.124 vs. 1.650 ± 0.097). This variance can also be determined using the ratio of uptake or S/B ratio between 68 Ga-FAPI-04 and 18 F-FDG as shown in Fig. 4c and d, with the ratio increasing gradually from week 3 to week 5, but dropping abruptly from week 5 to week 6.

Correlation between 18 F-FDG and 68 Ga-FPAI-04 uptakes
In order to fully characterize the relationship between the uptakes of 18 F-FDG and 68 Ga-FPAI-04, further correlation analysis was also performed with radioligand uptakes and S/B ratios during this dynamic process using the same VOIs. As shown in Fig. 5a, we did not observe a close correlation between the uptakes of 18 F-FDG and 68 Ga-FPAI-04 at week 5 or week 6. Even more, there is also no close correlation between the uptakes of 68 Ga-FAPI-04 at week 5 and the uptakes of 18 F-FDG at week 6. Similarly, no close correlation was also observed between the S/B ratio of these two radiotracers either at week 5 or week 6, as shown in Fig. 5b.

Comparison of 68 Ga-FPAI-04 and 18 F-FDG in lymph node pre-metastasis
The high uptake of 68 Ga-FAPI-04 (%ID/cc = 4.0) (n = 2) was also observed in the lymph node as shown in Fig. 6a. Interestingly, this lymph node does not have significant 18 F-FDG as shown in Fig. 6a. The resected tissue in Fig. 6b demonstrates a swelling lymph node with the same location detected by 68 Ga-FAPI-04 imaging. Further in vitro H&E staining of the swelling tissue also demonstrated an elevated FAP expression surrounding the lymph node as shown in Fig. 6c. However, in the H&E staining, we did not observe a remarkable tumor metastasis and a significant accumulation of cancer cells.

Discussion
This study focuses on the longitudinal imaging of preclinical 4T1 breast cancer model using 18 F-FDG and 68 Ga-FAPI-04. The 4T1 murine breast cancer cells derived from breast cancer cells in BALB/c mice. It can spontaneously metastasize from the primary tumor in the mammary gland to multiple distant sites including lymph nodes, blood, liver, lung, brain, and bone [27,28], normally 5 weeks after the 4T1 tumor cell implantation which was also monitored by H&E staining in this study (data not shown). In this study, we observed the metastatic lesion in both lung and lymph nodes with 68 Ga-FAPI-04 imaging. For the lung metastasis, we observed that 68 Ga-FAPI-04 holds an advantage over 18 F-FDG in detecting the metastasis at the early stage (week 5) for 18 F-FDG instead of 68 Ga-FAPI-04. Furthermore, in vitro H&E and IHC analysis of the metastatic lesions at different time points also demonstrates the expression of FAP change dynamically from week 5 to week 6. This dynamic change of FAP expression can be monitored using 68 Ga-FAPI-04 imaging, demonstrating a high uptake and S/B ratio at week 5 instead of week 6. At the early stage of metastasis, activated CAFs cells expressing FAP may contribute to the formulation of metastatic microenvironment, which lays the foundation for the development of tumor metastasis. However, with the progress of the tumor metastasis, the contents of microenvironment and expression of FAP may change dramatically. This dynamic change will indeed affect the detection rate of 68 Ga-FAPI-04 imaging in clinic, as demonstrated F-FDG during the longitudinal study. All %ID/cc was presented as the mean uptake. The same VOIs were employed for the %ID/cc and S/B analysis in this study. With this result, we envision that in tumor metastasis to the so-called "soil" [29], raised CAFs cells can help establish the early metastatic microenvironment. During this stage, 68 Ga-FAPI-04 is superior to 18 F-FDG in detecting the metastatic lesion. However, after the metastasis progresses to a certain stage, FAP expression drops and the metastatic lesion is mainly composed of tumor cells. At this time point, metastatic foci may show higher 18 F-FDG uptake and lower 68 Ga-FAPI-04 uptake, with 18 F-FDG demonstrating higher detection rate than 68 Ga-FAPI-04. Interestingly, further analysis did not show a strong correlation between the uptakes of 18 F-FDG and 68 Ga-FAPI-04. This reminds us that both of these two imaging modalities may provide different information on tumor metastasis and should be taken into account independently.
In this study, we also observed a high level of 68 Ga-FAPI-04 uptake and FAP expression for lymph node metastasis. The pre-metastatic induction of lymphangiogenesis has been well documented for many tumor entities [30][31][32]. Various evidences also indicate that premetastatic conditioning of the lymph node microenvironment recruits MDSCs, TAMs, and immature DCs to drive the growth of metastasis function as building immunosuppression microenvironment [33].Yeung et al. reported that myofibroblasts, which features high expression of FAP and α-SMA [34], will be substantially activated when colorectal micro-metastases develop within lymph nodes. However, once the metastatic lesions have reached a certain mass threshold, myofibroblasts return to a lower level of proliferation characterized by the low level of Ki67 expression [35]. In this study, we observed a high uptake of 68 Ga-FAPI-04 while tiny uptake of 18 F-FDG at the early stage of lymph node metastasis. This reminds us that we cannot simply assume 68 Ga-FAPI-04 imaging to be false positive when we find no 18 F-FDG uptake or significant number of cancer cells in histopathological sections. This high uptake of 68 Ga-FAPI-04 and low uptake of 18 F-FDG may indicate an early stage of metastasis which should be seriously monitored during the follow-up visits for the patients.
To a certain extent, 68 Ga-FAPI-04 imaging can detect early lung metastasis and lymph node metastasis of cancer. Nevertheless, the expression of FAP descends at the later growth stage of metastatic lesions, which could herald the limitation of FAPI  Tissue collected for the lymph node metastasis. c H&E and anti-FAP IHC staining for the corresponding lymph node metastasis in the diagnosis of tumor metastasis and tumor staging. In this study, we envision that 68 Ga-FAPI-04 may have an optimal time-window in visualizing the metastatic lesions. However, this study was only performed with one tumor model. Further preclinical studies with different tumor cell lines and tumor models still need to be performed to fully clarify this phenomenon in the future. Moreover, further longitudinal studies with different tumor types still need to be performed to fully characterize this dynamic process in clinic.
In this study, we studied a highly aggressive murine breast cancer cell (4T1 cell line) with the metastatic process progressed only in 1 to 2 weeks. However, in clinical practice, various types of tumors have different growth patterns and invasion rates, with metastasis often lasting from months to years. Considering this difference, it does not indicate human FAP expression is also as transient as the 4T1 cells. Furthermore, human astrocytoma, sarcoma, melanoma cell lines can also express FAP, which can also explain why we still see the 68 Ga-FAPI uptake in some of metastasis in human.
In summary, 68 Ga-FAPI-04 is more sensitive to 18 F-FDG to detect the early stage of tumor metastasis, but becomes less sensitive to 18 F-FDG at the later stage of tumor metastasis. The dynamic change of 68 Ga-FPAI-04 will change the detection rate of tumor metastasis and following tumor staging. We envision this result would be meaningful for the explanation of the 18 Ga-FAPI-04 and 18 F-FDG imaging both in the future clinic and preclinic studies.